An Integrated Multi-Criteria Decision Analysis and Structural Equation Modeling Application for the Attributes Influencing the Customer’s Satisfaction and Trust in E-Commerce Applications

Abstract

Since the COVID-19 pandemic has forced most industries to embrace an online platform utilizing technological breakthroughs, it has significantly impacted our daily lives. Businesses that use marketplaces to sell and trade products to customers while increasing their participation through online shopping or e-commerce are among the sectors that take advantage of these situations. The current study set out to evaluate the level of customer satisfaction, trust, and service quality of the e-commerce application to enhance the system and provide a better shopping experience. Facebook, Shopee, Lazada, Shein, and TikTok were the five e-commerce platforms evaluated. The Philippines was the location of this study, and at least 200 people answered the survey, which was conducted in-person and online and consisted of 72 questionnaires. The researchers assessed twelve latent variables: perceived security, customer satisfaction, application interface, brand equity, tangibility, reliability, responsiveness, assurance, empathy, and information credibility. Structural equation modeling and multi-criteria decision analysis were used to analyze the data. The findings demonstrated that assurance, application interface, information credibility, and brand equity directly impacted service quality. Moreover, a direct and significant correlation exists between customer satisfaction and service quality. Customers’ trust is significantly impacted by their level of satisfaction and perception of security. The e-commerce apps were ranked using a multi-criteria decision analysis technique, which is the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) based on tangibility, responsiveness, assurance, reliability, and empathy. Facebook received a low rating, whereas Shopee was ranked highest. To further enhance the services offered, the lowest rated application may use the results from the combination of the TOPSIS and SEM results. Additionally, application providers, managers, and researchers examining the user–software interaction of relevant e-commerce applications might utilize the study’s results to enhance their services regarding the purchasing experience and provide a sustainable service. Finally, this study is among the first to use the structural modeling approach to evaluate customer trust and satisfaction while integrating service quality and TOPSIS.

1. Introduction

The continuous growth of technology and the evolution of the Internet have led to the emergence of various significant revolutions in platforms to improve daily life activities, including shopping [1]. Hence, marketplace businesses take advantage of these instances to sell and trade products to customers while expanding their engagement through online shopping, also known as e-commerce. Through e-commerce, customers attain convenience in buying and conducting payments, access to a wide range of products, and the capacity to compare prices and products from multiple retailers using various online channels and mediums [2]. These result in an increase in competition in the marketplace, so new opportunities like the use of mobile apps to provide direct services for better relationship management and communication with consumers through smartphones have been adopted [3].

E-commerce mobile apps are applications that can be installed on smartphones and are ubiquitous tools that people can carry with them [4]. Users can conveniently shop online, browse for a variety of products that are not available in local markets, create wish lists, add items to their cart, and complete purchases [5] at any time and location as they have access to the Internet by registering for an account and providing their e-mail, contact number, and home address [6] to deliver the ordered product in their preferred delivery options [7], and just wait for their order to be delivered to their homes [8]. In addition, it also has multiple payment options for consumers, such as cash-on-delivery, credit/debit card, e-wallet, and other methods that facilitate and streamline the checkout process efficiently [9]. Furthermore, the e-commerce mobile app sends direct messages to clients via push notifications, informing them of new items, reminding them of their browser activities and items in their shopping cart, and providing updates on the location and status of the purchased goods. Moreover, it has additional features, including games and rewards programs that give discounts and vouchers to customers [10], a live chat feature that helps retailers and customers better communicate about prices and product conditions [11], and open comment boxes within product pages where customers leave their issues and ratings [12] that other customers can assess to evaluate whether they will be satisfied with their purchase, as well as provide feedback about what retailers should improve in their business.

However, the COVID-19 pandemic crisis in 2020 became a push factor for people to explore online shopping [13]. The established health protocols and lockdown restrictions have led to an increase in e-commerce demands from customers and opened an innovation opportunity for businesses to engage in e-commerce-structured business in the marketplace. In the Philippines, as of January 2021, the International Trade Administration reported that the country had about 73 million active online users, with online shopping activities increasing to 23%, and most are accessed on mobile devices [10]. However, despite the growing popularity among the general public and the application of new methods, such as the development of e-commerce mobile apps to reach and engage more consumers in online shopping, e-commerce still faces many challenges in online shopping, affecting both online retailers and online buyers [4].

A study by [14] shows a four-dimensional construct of e-service quality: website design, customer service, security and privacy, and fulfillment. Customer satisfaction reflects how well the goods fulfill the buyer’s needs and preferences [13]. It also indicates the effectiveness of the business [13] and influences the willingness of customers to perform online purchases of goods in repeat [15].

According to [16], the decreased presence of human and social elements in the online environment impairs a trustworthy and sociable online transaction environment [16]. As the usage of e-commerce platforms grows, they also become more complex and diverse in their online interactions and vulnerable to fraud, hacking, and digital theft [17]. Many users hesitate to provide personal data online because of doubts about e-commerce security [17] concerning privacy data abuse and illegal accessibility of transaction records [18]. As stated by [17], trust has a crucial impact on the customer’s behavior and e-commerce success. Moreover, Cheng et al. distinguished trust into two types: trust towards social commerce members and trust towards social commerce apps. The former discusses the provision of reliable information provided by retailers and customers’ reviews, and the latter reviews the design layout, ease of use, and interface system [18]. This study integrated a Service Quality Model (SERVQUAL), which is a tool used to compare the various organization’s service quality performance to customer service quality needs [19]. It aims to explore and analyze the trustworthiness of e-commerce mobile apps based on customer preferences. In addition, the model constructs in this study can be used to scrutinize the key variables identified as having a considerable impact on the behavior of the customer to use the e-commerce apps as a means of platform for online shopping and figure out how to make it more convenient and utilized for the end-users (customers and retailers). The objective of the study is to analyze the significant factors impacting consumer preferences for e-commerce applications in the context of trust, to create a structural equation model to assess the correlations between consumer trust and latent variables, and to apply multi-criteria decision analysis Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS) techniques to determine the highest and lowest level of trustworthy e-commerce application among the alternatives through multiple criteria.

This current study has demonstrated substantial advancements in the field of e-commerce, particularly concerning enhancing service quality, minimizing operational expenses, and evaluating customer preferences and satisfaction. This particular research aims to provide a comprehensive analysis of the primary aspects that influence customers’ trust in e-commerce applications. This information is crucial for e-commerce businesses to address trust-related concerns faced by their clients and implement strategies to boost trustworthiness. Moreover, the application of the combined Multi-Criteria Dimension Analysis (MCDA) method, namely TOPSIS and Service Quality (SERVQUAL) model, to assess the service quality of e-commerce mobile apps in the Philippines in terms of tangibility, responsiveness, reliability, assurance, and empathy that influences customer satisfaction bridging to the built of customer’s trust, that has not been applied in other studies.

Hypothesis Development and Related Literature

The conceptual framework in Figure 1 of the study integrates Multi-Criteria Decision Analysis (MCDA) to evaluate customer preferences in e-commerce applications. The framework has 12 latent variables: responsiveness, reliability, assurance, empathy, brand equity, application interface, information credibility, service quality, customer satisfaction, security, and trust. The study aims to assess the relationship between these variables and their impact on customer behavior and e-commerce success. The framework also incorporates the Service Quality Model (SERVQUAL) is a multidisciplinary research instrument that aims to gather consumer expectations and perceptions about the services across five dimensions, which are thought to constitute service quality and the Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS) to compare the service quality performance of e-commerce applications and determine the highest and lowest levels of trustworthiness among the alternatives [20]. This framework provides a clear approach to understanding and analyzing the factors affecting consumer preferences for e-commerce applications and their trust in digital marketplaces.

Hypothesis 1 (H1). 

There is a significant relationship between tangibility and service quality.

In terms of tangibility as a factor in reducing feelings of anxiety, developers must consider customers’ wants and desires in terms of design, as usability is the starting point for gaining users’ trust and support for the site. The tangibility of the online presence aids in reducing heterogeneity in the execution of services and, as a result, improves quality control [21].

Hypothesis 2 (H2). 

There is a significant relationship between responsiveness and service quality.

Responsiveness refers to the e-commerce application shop’s readiness for prompt service. It is important to respond to all user’s or customers’ needs and inquiries in a courteous way; otherwise, it will become a complaint about the e-commerce application’s service quality, which can result in low customer satisfaction [22]. According to the study conducted by [23], this dimension is positively associated with the user’s intention to repurchase, which increases customer trust.

Hypothesis 3 (H3). 

There is a significant relationship between reliability and service quality.

Reliability reflects the honesty and accuracy of the information provided by the e-commerce application [24]. The consistency and integrity of the service provider in executing the delivery over time without any interruptions or errors can measure the precision and reliability of the information. This dimension is one of the significant factors that affect service quality, as it depicts the credibility of the description, pricing policy, and service delivery of the products purchased in an e-commerce application [25].

Hypothesis 4 (H4). 

There is a significant relationship between assurance and service quality.

Assurance indicates the ability of the e-commerce application to inspire trust and confidence through the accuracy of the information provided by the retail online shops and the knowledge and courtesy of its employees [24]. The users should feel safe and secure during and after the transaction in terms of the privacy and security of their personal information and the assurance of their payment [26]. Seven dimensions were originally developed for assurance and empathy, which are credibility, courtesy, security, privacy, competence, and understanding or knowing users’ or customer’s needs [22].

Hypothesis 5 (H5). 

There is a significant relationship between empathy and service quality.

Empathy is an expression of the right communication skills and job knowledge while offering related services. The ability of a staff member to communicate well, understand customers, and offer individual attention to guests is essential, as it promotes customer satisfaction and customer trust [27]. It is important for e-commerce shop employees to meet their customers’ needs and understand their requests and inquiries so that this builds up a good relationship between the customer and the employee.

Hypothesis 6 (H6). 

There is a significant relationship between brand equity and service quality.

The brand plays a crucial role in achieving business continuity. The perception of a brand that reflects the consumer’s recall of its association with the brand is referred to as a brand image. According to the findings of this study, the store brand image and service quality have a significant impact on customer satisfaction, trust, and commitment to becoming loyal customers [28].

Hypothesis 7 (H7). 

There is a significant relationship between application interface and service quality.

Numerous studies have established a significant and influential relationship between the quality of an application interface and overall service quality perception [29,30]. The design and presentation of an application play a role in shaping users’ perception of the service quality. It also has an impact on their satisfaction with the experience they have while using the service.

Hypothesis 8 (H8). 

There is a significant relationship between information credibility and service quality.

In some empirical studies, credibility as a source of information appears to be the biggest factor affecting e-commerce quality. In the virtual environment, it is vital to gain customer satisfaction and trust by performing what it promises to do, which includes the accuracy of delivery service, complete order service, being truthful about the offerings, and having the website always ready and available. Moreover, credibility can be shown through the consistency, reliability, and accuracy of the information presented by the e-commerce application and its shop [31].

Hypothesis 9 (H9). 

There is a significant relationship between service quality and customer satisfaction.

Extensive research has consistently demonstrated a significant and positive relationship between service quality and customer satisfaction [20]. Companies that place importance on and improve the caliber of their offerings are more likely to see increased levels of customer contentment as the perceived excellence of services has an impact on customer satisfaction. In the study of [32], it is emphasized that service quality has a direct influence on customer satisfaction.

Hypothesis 10 (H10). 

There is a significant relationship between customer satisfaction and customer trust.

The customer’s selection has a substantial impact on their satisfaction with the product or service and the customer’s trust [33]. The customer’s repurchase intention is considerably affected if they are satisfied with the product/service they acquired online [34]. This indicates customers are satisfied with a company’s products or services, and they are more likely to trust the business and its intentions. This trust can be further strengthened by consistently meeting or exceeding customer expectations, resulting in a positive cycle of trust and satisfaction.

Hypothesis 11 (H11). 

There is a significant relationship between security and customer trust.

Trust is the key to success for successful online businesses in a thriving e-commerce environment. Consumers are more sensitive to the safety of their Internet now that technology has advanced so rapidly. Therefore, e-shopping companies that are striving to maintain customer loyalty and increase their sales must first address safety concerns and build trust [35].

2. Methodology

2.1. Participants

A mixed-mode survey was used in this study, in which researchers used both face-to-face interaction and an online survey to gather data. Researchers disseminated printed survey questionnaires for face-to-face interaction, and online questionnaires via link survey were sent using Messenger and other social media platforms. A total of 235 respondents answered the survey. They were individuals who had experience ordering goods from online shops using different e-commerce applications. The study of [36] states that a study with 10 to 15 indicators should fall between 200 and 400 sample sizes for structural equation models. Thus, the gathered statistics have been considered acceptable.

2.2. Questionnaire

The questionnaire was designed by the researchers to investigate and assess the customer’s preference in choosing and using a trustworthy e-commerce application. The questionnaire is divided into twelve (12) latent variables. The survey questionnaire consists of 72 questions with twelve sections (

Table 1. The constructs and measurement items.

Construct	Items	Measures	Supporting Reference
Tangibility (T)	T1	The e-commerce application provides a clear product presentation, making the product easy to comprehend.	[37]
	T2	The e-commerce application presents products in a tangible manner (images, descriptions, etc.)	[38]
	T3	I believe that I could successfully conduct any financial transactions using the E-wallet.	[39]
	T4	The layout of the e-commerce application is clutter-free and professionally designed.	[40]
	T5	The navigation system of the application is easy to learn.	[21]
	T6	The e-commerce application gives the perception of how real the interaction with products feels.	[41]
Responsiveness (RP)	RP1	The e-commerce application provides complete and detailed information before the user’s need for the products.	[42]
	RP2	The e-commerce application indicates the estimated date and time of delivery.	[43]
	RP3	The e-commerce application gives clear, understandable information.	[44]
	RP4	The e-commerce application is willing to accept criticism from customers using a star rating in the Play Store.	[45]
	RP5	The e-commerce application is willing to accept advice from customers through a comment box in the Play Store.	[45]
	RP6	The e-commerce application informs customers about the condition and location of goods and products that they order.	[46]
Reliability (RL)	RL1	The e-commerce application provides accurate information.	[42]
	RL2	The logistics company met the estimated date and time indicated by the e-commerce application.	[43]
	RL3	The service of the e-commerce applications is delivered with accuracy.	[47]
	RL4	The security of the mode of payment on the e-commerce application can be trusted.	[48]
	RL5	The e-commerce application, in general, understands customers who need services, especially when an immediate response should be taken.	[49]
	RL6	The privacy of the customer’s personal information on the e-commerce application can be trusted.	[48]
Assurance (A)	A1	The e-commerce application (Shopee, Lazada, TikTok, etc.) provides a secure transaction in purchasing goods online.	[42]
	A2	Retail shops with e-commerce applications know how to answer your questions.	[43]
	A3	Retail shops are trustworthy in e-commerce applications.	[50]
	A4	E-commerce retail shops are consistently courteous in interacting with their customer.	[51]
	A5	I feel secure and confident using any method of payment (e-wallet, COD, credit card, etc.) when purchasing goods online using an e-commerce application.	[52]
	A6	I feel safe doing my transactions in purchasing goods online using an e-commerce application.	[52]
Empathy (EM)	EM1	E-commerce shops communicate with the customer sincerely.	[53]
	EM2	E-commerce shops respond to the customer’s complaints properly.	[53]
	EM3	E-commerce shops are helpful, careful, and friendly.	[44]
	EM4	E-commerce shops have convenient business hours.	[54]
	EM5	The e-commerce shop listens to and actively addresses customers’ needs and concerns.	[55]
	EM6	Overall, e-commerce shops are respectful and optimistic when approaching customers.	[44]
Brand Equity (BE)	BE1	The e-commerce application provides convenience during online shopping.	[56]
	BE2	I am aware of various e-commerce applications that can be used for online shopping.	[57]
	BE3	The E-commerce application provides a positive environment for my online shopping.	[57]
	BE4	An e-commerce application has a Live Chat feature that can help me to negotiate prices directly.	[11]
	BE5	I can recognize an e-commerce application just by looking at the design and appearance of the application.	[11]
Application Interface (AI)	AI1	The e-commerce application is user-friendly (intuitive, easy to use, and simple).	[58]
	AI2	The e-commerce application provides a utilized navigation search bar that helps me find my top products with ease.	[58]
	AI3	The application interface is visually appealing.	[59]
	AI4	The existing product reviews and ratings are displayed and organized on the product pages.	[12]
	AI5	The e-commerce application loads information quickly.	[60]
	AI6	I can easily find the information that I am looking for using the e-commerce application.	[61]
	AI7	The online shopping application is easy to navigate using the e-commerce application.	[62]
Information Credibility (IC)	IC1	I believe the product descriptions on the e-commerce app are accurate and reliable.	[63]
	IC2	I trust an e-commerce app’s reliability in delivering accurate and up-to-date product information.	[63]
	IC3	I trust the customer reviews on the e-commerce application.	[63]
	IC4	I consider evaluating the credibility of information by taking into account the e-commerce app’s privacy and security policies.	[64]
	IC5	I regularly check out the information in an e-commerce app before purchasing a product.	[64]
	IC6	I believe the trustworthiness of the e-commerce application is enhanced by the accuracy and transparency of the product information.	[64]
Service Quality (SQ)	SQ1	It is easy to find the products you are looking for on the app.	[65]
	SQ2	The product descriptions on the app were accurate and helpful.	[66]
	SQ3	The customer service team was responsive and helpful to my inquiries	[67]
	SQ4	It was easy to add items to my cart and checkout on the e-commerce app.	[68]
	SQ5	The checkout process was fast and secure.	[68]
	SQ6	The e-commerce app has a secure payment process.	[68]
Customer Satisfaction (CS)	CS1	Overall, I am satisfied with using e-commerce applications to make online purchases.	[69]
	CS2	The application system knows its users well enough to offer them products and services adapted to the consumer’s needs based on their preferences and market conditions.	[70]
	CS3	I feel satisfied with the product I usually buy through an e-commerce application.	[71]
	CS4	I am satisfied with my experience whenever I buy a product through an e-commerce application.	[71]
	CS5	I am confident in purchasing products through the use of e-commerce applications.	[72]
	CS6	I am satisfied with the information and services provided.	[72]
Perceived Security (PS)	PS1	I feel safe providing sensitive information about myself over the e-commerce applications.	[73]
	PS2	I am confident that my personal information is kept confidential and safe when using e-commerce applications.	[13]
	PS3	I believe that my financial-related information is protected from unauthorized access.	[18]
	PS4	I perceive that e-commerce applications are secured systems to conduct a transaction.	[73]
	PS5	I feel secure using various payment channels (cash on delivery, credit/debit card, online banking, etc.) through e-commerce applications.	[73]
	PS6	It has sufficient technical capacity to ensure that the data I send cannot be modified by hackers.	[70]
Customer’s Trust (TR)	TR1	I trust the e-commerce application as a safe system for doing online shopping.	[74]
	TR2	It is safe to pay money and perform a financial transaction using an e-commerce application.	[75]
	TR3	I trust that the e-commerce sellers will not display my personal information (e-mail, phone number, name...) to others for commercial use.	[75]
	TR4	I am confident that the application system will resolve any issues or disputes related to my transactions.	[76]
	TR5	I trust the system characteristics of the e-commerce application.	[77]
	TR6	The e-commerce sellers provide reliable information for my online purchasing and payment transactions.	[78]

): tangibility (1), responsiveness (2), reliability (3), assurance (4), empathy (5), brand equity (6), application interface (7), information credibility (8), service quality (9), customer satisfaction (10), security (11), and customer’s trust (12). All the questions have been carefully formulated to help researchers understand the customer’s preference for the quality of services of an e-commerce application and how this will affect user trust. This study used Likert-type scales with numerical values (1 = strongly disagree, 2 = disagree, 3 = neutral, 4 = agree, and 5 = strongly agree) to measure the responses.

2.3. Structural Equation Modeling

SEM is a method used for investigating and analyzing multimodal relationships between latent variables and is suitable for examining a set of hypotheses from an entire model, combining regression and factor analysis [79]. Moreover, it is a multivariate technique that is commonly used for testing and simulating model hypotheses in the areas of social and behavioral science [80]. In this study, SEM tests twelve (12) latent variables: Tangibility, Responsiveness, Reliability, Assurance, Empathy, Brand Equity, Application Interface, Information Credibility, Service Quality, Customer Satisfaction, Security, and Customer Trust.

3. Results

3.1. Structural Equation Modeling Results

Figure 2 demonstrates the initial SEM for the factors influencing the service quality and trustworthiness of the e-commerce applications. According to the figure below, three hypotheses were not significant: responsiveness to service quality (Hypothesis 2), reliability to service quality (Hypothesis 3), and empathy to service quality (Hypothesis 5). Therefore, a revised SEM was derived by removing these hypotheses. The researchers modified some indices to enhance the model fit based on previous studies using the SEM approach. Figure 3 shows the final SEM for investigating the variables influencing the service quality and trustworthiness of e-commerce applications.

Table 2. Summary of hypotheses.

Hypothesis		p-Value	Interpretation
H1	There is a significant relationship between tangibility and service quality.	0.033	Significant
H2	There is a significant relationship between responsiveness and service quality.	0.324	Not Significant
H3	There is a significant relationship between reliability and service quality.	0.074	Not Significant
H4	There is a significant relationship between assurance and service quality.	0.017	Significant
H5	There is a significant relationship between empathy and service quality.	0.234	Not Significant
H6	There is a significant relationship between brand equity and service quality.	0.013	Significant
H7	There is a significant relationship between application interface and service quality.	0.002	Significant
H8	There is a significant relationship between information credibility and service quality.	0.001	Significant
H9	There is a significant relationship between service quality and customer satisfaction.	0.001	Significant
H10	There is a significant relationship between consumer satisfaction and customer trust.	0.002	Significant
H11	There is a significant relationship between security and customer trust.	0.001	Significant

presents that tangibility has a significant relationship with service quality, with a p-value of 0.033; responsiveness has no significant relationship with service quality, with a p-value of 0.324; reliability has no significant relationship with service quality, with a p-value of 0.074; assurance has a significant relationship with service quality, with a p-value of 0.017; and empathy has no significant relationship with service quality, with a p-value of 0.234. On the other hand, the remaining latent variables show a significant relationship to service quality with a p-value less than 0.05. Moreover, service quality has a significant relationship with customer satisfaction, with a p-value of 0.001, and customer satisfaction has a significant relationship with customer trust, with a p-value of 0.002. With a p-value of 0.001, perceived security has a significant relationship with customers’ trust.

Table 3. Descriptive statistics result.

				Factor Loading
Factor	Item	Mean	SD	Initial Model	Final Model
Tangibility	T1	3.8596	0.77263	0.717	0.728
	T2	3.9254	0.76179	0.735	0.723
	T3	3.8377	0.77686	0.743	0.73
	T4	3.7412	0.75874	0.69	0.711
	T5	3.9561	0.75874	0.723	0.743
	T6	3.7325	0.75874	0.721	0.741
Responsiveness	RP1	3.7982	0.77607	0.729	-
	RP2	3.9518	0.75813	0.709	-
	RP3	3.9123	0.78039	0.81	-
	RP4	3.7939	0.85852	0.651	-
	RP5	3.8553	0.79154	0.661	-
	RP6	3.9912	0.76251	0.622	-
Reliability	RL1	3.7588	0.70783	0.811	-
	RL2	3.7456	0.71252	0.671	-
	RL3	3.7588	0.78458	0.74	-
	RL4	3.8289	0.71586	0.591	-
	RL5	3.7632	0.76028	0.686	-
	RL6	3.8026	0.8346	0.686	-
Assurance	A1	3.7544	0.75748	0.715	0.747
	A2	3.7807	0.72414	0.671	0.619
	A3	3.7325	0.75291	0.711	0.692
	A4	3.7763	0.73787	0.71	0.678
	A5	3.7675	0.81473	0.692	0.703
	A6	3.8246	0.69289	0.749	0.755
Empathy	EM1	3.7588	0.66283	0.73	-
	EM2	3.7325	0.75874	0.704	-
	EM3	3.7588	0.79017	0.736	-
	EM4	3.8377	0.75383	0.666	-
	EM5	3.7895	0.73878	0.718	-
	EM6	3.8377	0.65368	0.731	-
Brand Equity	BE1	3.9386	0.71836	0.686	0.699
	BE2	3.886	0.75977	0.729	0.741
	BE3	3.8333	0.67578	0.709	0.68
	BE4	3.8421	0.80832	0.617	0.566
	BE5	3.9298	0.77652	0.672	0.682
Application Interface	AI1	3.9868	0.72392	0.732	0.702
	AI2	4.0175	0.72077	0.688	0.691
	AI3	3.9693	0.6986	0.716	0.712
	AI4	3.9342	0.68339	0.719	0.705
	AI5	3.7149	0.74049	0.578	0.577
	AI6	3.8553	0.67106	0.614	0.626
	AI7	3.8904	0.69059	0.799	0.78
Information Credibility	IC1	3.6886	0.77086	0.568	0.527
	IC2	3.807	0.68819	0.614	0.585
	IC3	3.8246	0.75379	0.725	0.683
	IC4	3.9211	0.73484	0.745	0.765
	IC5	4.0395	0.80375	0.671	0.714
	IC6	3.9825	0.7022	0.731	0.723
Service Quality	SQ1	3.8947	0.73152	0.495	0.528
	SQ2	3.8684	0.69629	0.541	0.572
	SQ3	3.7675	0.74703	0.436	-
	SQ4	4.0175	0.70844	0.515	0.576
	SQ5	3.9737	0.72659	0.546	0.595
	SQ6	3.8202	0.74392	0.486	0.5
Customer Satisfaction	CS1	3.9035	0.71447	0.491	0.472
	CS2	3.9167	0.6879	0.456	-
	CS3	3.9386	0.6474	0.528	0.45
	CS4	3.9035	0.70203	0.462	-
	CS5	3.8377	0.68655	0.471	-
	CS6	3.8947	0.6686	0.562	0.485
Perceived Security	PS1	3.6096	0.80816	0.726	0.692
	PS2	3.7237	0.73188	0.803	0.773
	PS3	3.6711	0.80288	0.695	0.698
	PS4	3.7851	0.71012	0.758	0.739
	PS5	3.7632	0.7062	0.738	0.767
	PS6	3.7851	0.76966	0.786	0.818
Customer’s Trust	TR1	3.8114	0.6926	0.705	0.687
	TR2	3.7281	0.69339	0.621	0.617
	TR3	3.7763	0.77855	0.597	0.615
	TR4	3.7675	0.74111	0.593	0.612
	TR5	3.7939	0.68112	0.587	0.598
	TR6	3.8202	0.7075	0.644	0.636

displays and explains the statistical outcomes of the evaluated variables. Every factor loading in the Initial Structural Equation Model (SEM) that is equal to or greater than 0.50 denotes reliability and supports the validity of the model [81]. Thus, the factors loading that are less than 0.50 were removed, and the loadings under the insignificant indicators were also not included in the evaluation of the Final SEM.

Table 4. Model validity.

Factor	Cronbach’s α	Average Variance Extracted (AVE)	Composite Reliability
Tangibility	0.866	0.5208	0.9747
Responsiveness	0.851	0.4896	0.9739
Reliability	0.849	0.491	0.9808
Assurance	0.856	0.5018	0.9736
Empathy	0.86	0.5016	0.9752
Brand Equity	0.811	0.4674	0.9565
Application Interface	0.865	0.4608	0.9604
Information Credibility	0.832	0.484	0.9806
Service Quality	0.86	0.2853	0.9839
Customer Satisfaction	0.863	0.2973	0.8754
Perceived Security	0.884	0.5653	0.9772
Customer’s Trust	0.892	0.3917	0.9654

shows that Cronbach’s alpha of twelve latent variables is more significant than 0.7. The Average Variance Extracted (AVE) from tangibility to other latent variables, such as Responsiveness, Reliability, Assurance, Empathy, Brand Equity, Application Interface, Information Credibility, and Perceived Security, are all greater than 0.4, and the composite reliability of all nine latent variables is more significant than 0.8. However, Service Quality, Customer Satisfaction, and Customer Trust are less than 0.5. The value of AVE for every construct should be higher than 0.5; however, if the value of AVE is less than 0.5 but the composite reliability is greater than 0.6, the convergent validity of the construct is acceptable [82].

According to [83], for structural equation models, at a minimum, the following indices should be shown: the model RMSEA, the CFI, and the SRMR. The RMSEA value was 0.088, which is less than the recommended value, depicting the marginal value, according to [84], as shown in

Table 5. Model fit.

Goodness-of-Fit Measures of SEM	Parameter Estimates	Minimum Cutoff	Reference
Root Mean Square (RMSEA)	0.088	≤0.10	[84]
Comparative Fit Index (CFI)	0.729	>0.70	[85]
Standardized RMR	0.3417	<0.08	[86]

. According to [85], the CFI value was 0.729, which is more significant than the suggested reduction of 0.70. The SRMR value was 0.3417, higher than the minimum threshold, as defined by [86].

Table 6. Direct, indirect, and total effects.

No	Variables	Direct Effects	p-Value	Indirect Effects	p-Value	Total Effects	p-Value
1	AI -> SQ	0.513	0.002	-	-	0.513	0.002
2	IC -> SQ	0.577	0.001	-	-	0.577	0.001
3	BE -> SQ	0.246	0.023	-	-	0.246	0.023
4	A -> SQ	0.428	0.001	-	-	0.428	0.001
5	T -> SQ	0.129	0.057	-	-	0.129	0.057
6	AI -> CS	-	-	0.607	0.002	0.607	0.002
7	IC -> CS	-	-	0.684	0.001	0.684	0.001
8	BE -> CS	-	-	0.292	0.022	0.292	0.022
9	A -> CS	-	-	0.507	0.001	0.507	0.001
10	T -> CS	-	-	0.153	0.054	0.153	0.054
11	SQ -> CS	1.184	0.001	-	-	1.184	0.001
12	PS -> CS	-	-	-	-	-	-
13	TR -> CS	-	-	-	-	-	-
14	AI -> TR	-	-	0.223	0.002	0.223	0.002
15	IC -> TR		-	0.251	0.001	0.251	0.001
16	BE -> TR	-	-	0.107	0.016	0.107	0.016
17	A -> TR	-	-	0.186	0.001	0.186	0.001
18	T -> TR	-	-	0.056	0.049	0.056	0.049
19	SQ -> TR	-	-	0.436	0.002	0.436	0.002
20	PS -> TR	0.851	0.001	-	-	0.851	0.001
21	CS -> TR	0.368	0.002	-	-	0.368	0.002

indicates the correlations among the variables. This table shows that all variables have a significant direct effect and total effect with a p-value of less than 0.05, except the tangibility to service quality with a p-value of 0.057. Moreover, all the relationships have significant indirect effects with also a p-value less than 0.05, except the tangibility to customer satisfaction with a p-value of 0.054.

3.2. Technique for Order Preference by Similarity to an Ideal Solution Result

The Technique for Order Preference by Similarity to an Ideal Solution is one of the well-known multiple criteria decision analyses proposed by [87]. Its function is to identify the best alternative that is nearest to the positive ideal solution and farthest from the negative ideal solution [88]. In the current study, the TOPSIS is integrated with the SERVQUAL factors, including responsiveness, tangibility, reliability, assurance, and empathy, to investigate the trustworthiness of the five e-commerce applications based on customer preferences. Moreover, the five e-commerce applications that were observed are Shopee, Lazada, TikTok, Shein, and Facebook.

Table 7. Generation Z (aged 8–23 years old).

Alternatives	Dev from S+	Rank
Shopee	0.979007553	1
Lazada	0.173620998	4
Shein	0.471998483	3
TikTok	0.875068194	2
Facebook	0.117214239	5

shows that in Generation Z, the most trustworthy e-commerce application was Shopee, while the least was Facebook.

According to

Table 8. Millennials (aged 24–39 years old).

Alternatives	Dev from S+	Rank
Shopee	1	1
Lazada	0.901311428	2
Shein	0.584624786	4
TikTok	0.752030666	3
Facebook	0	5

, Shopee is the most trusted e-commerce app among Millennials, whereas Facebook is the least trusted.

In Generation X, Shopee is the most trusted e-commerce app, while Shein is the least trusted, as

Table 9. Generation X (aged 40–55 years old).

Alternatives	Dev from S+	Rank
Shopee	1	1
Lazada	0.383498590	4
Shein	0	5
TikTok	0.655838931	2
Facebook	0.515382493	3

indicates.

Table 10. Generation Z, Millennials, and Generation X.

Alternatives	Dev from S+	Rank
Shopee	1	1
Lazada	0.738966553	3
Shein	0.493349717	4
TikTok	0.744053916	2
Facebook	0	5

demonstrates that Shopee is the most trusted e-commerce app among 235 respondents, whereas Facebook is the least trusted.

According to the investigation, the most trustworthy e-commerce application overall was Shopee, followed by TikTok, Lazada, and Shein, while the least trustworthy e-commerce application was Facebook.

4. Discussion

The study focused on assessing the trustworthiness of e-commerce applications based on customer preferences. TOPSIS was used to rank the five e-commerce applications, namely Shopee (E1), Lazada (E2), Shein (E3), TikTok (E4), and Facebook (E5), with the use of the five dimensions: tangibility, responsiveness, reliability, assurance, and empathy. A total of 235 responses were acquired through an online questionnaire and face-to-face survey. The population is categorized into five (5) generations, including Gen Z, Millennials, Gen X, Pre-Boomers, and Boomers. However, only Gen Z, Millennials, and Gen X are the main respondents of the study. The e-commerce application ranking also demonstrates that E5 has the lowest ranking; to cope with E1, the provider must focus on improving the dimensions mentioned.

An SEM was utilized to analyze the correlations among responsiveness (RP), tangibility (T), reliability (RL), assurance (A), empathy (EM), and service quality (SQ), together with application interface (AI), brand equity (BE), information credibility (IC), customer satisfaction (CS), perceived security (PS), and trust (TR).

The findings of the current study indicated that service quality is directly affected by the following: assurance (β = 0.428, p = 0.001), application interface (β = 0.513, p = 0.002), information credibility (β = 0.577, p = 0.001), and brand equity (β = 0.246, p = 0.023). However, tangibility was shown to have a close direct effect on service quality (β = 0.129, p = 0.057). Thus, the providers must focus on improving the application interface, information credibility, brand equity, and assurance because these factors directly affect the service quality that bridged the growth of customer trust and intention in using the e-commerce application.

The security and comfort in using the service delivery that provides trust and confidence to the customer is referred to as assurance [89]. With the strength of this dimension, the customer can be assured that the service and products they purchased will be carried out safely and will meet the quality standards. In addition, according to [90], interface characteristics such as layout quality, navigation appearance, and visual appeal quality affected the overall trust of the customer. This factor meets the satisfaction of the customer to easily manage how to use and navigate the functions and services that they want to acquire in online shopping. Credibility defines the accuracy and provision of high-quality services, prices, safe providers, and well-presented instructions. The result of this research regarding the significance of credibility to service quality for customer satisfaction is affirmed by the study of [91]. Moreover, brand equity is one of the important factors that could define the service quality of a company business [92] as it takes a grasp of the customer’s awareness. The improvement in service quality establishes a positive appeal to the consumers to patronize the services provided by the company [93]. It is a competitive advantage to establish a unique design and service performance that will provide familiarity with a particular application and a positive environment to attain customer satisfaction and trust.

Ref. [94] obtained a result in their study that customer satisfaction directly affected the customer’s trust in e-commerce settings, which is also emphasized in the current study (β = 0.368, p = 0.002). Lastly, the results show that perceived security directly affected the customer’s trust in using the e-commerce application (β = 0.851, p = 0.001). It is indicated in the study of [95] that the perception of security focuses on the provision of online privacy, protection of personal information, and prevention of unauthorized access to transactions, which have a significant and huge impact on trust.

5. Conclusions

During the COVID-19 pandemic, e-commerce experienced a sharp increase in popularity, leading to the creation of numerous e-commerce marketplaces and millions of downloads from various application stores. Consequently, cybercrime is a growing threat that may have a negative impact on customer satisfaction and trust. Thus, to examine the factors that directly and significantly impact service quality as it relates to pleasure and trust, which are also influenced by perceived security, the current study included a SERVQUAL model. Information credibility has the greatest beneficial impact on the service quality of e-commerce mobile applications (0.577), followed by application interface (0.513), assurance (0.428), and brand equity (0.246), according to the SEM result. Additionally, customer satisfaction is significantly correlated with service quality (1.184). This indicates that the level of performance of the services is directly correlated to the satisfaction of the customer. Moreover, customer satisfaction (0.368) and perceived security (0.851) have a strong and direct impact on trust.

To improve service quality, e-commerce mobile app providers ought to prioritize enhancing the aspects of assurance, brand equity, application interface, and information credibility, according to the study’s findings. Additionally, it enhances consumer satisfaction and security, which helps to build the consumer’s trust in the e-commerce application for their intended purchases.

The present study’s findings could potentially benefit users, developers, and scholars who are interested in studying unique applications of technology, like digitization. This research can be used as a basis for designing the most appropriate application features for their people, which could result in better service and reliable applications for our daily lives since digitalization and automation are some of the developments that most developing countries are currently adopting.

6. Limitations and Future Research

According to the 2020 Census of Population and Housing: Age and Sex Distribution in the Philippine Population, the Philippines has a total population of 109,035,343. The total population is divided into age brackets, signifying different generations. Based on the census, ages 5–24 (Generation Z) had a total population of 9,362,325 persons; ages 15–30 (Millennials) had a total population of 11,374,597 persons; ages 15–64 had a total population of 12,433,053 persons; and the total population of Occidental Mindoro was 525,354 persons [96]. Thus, this study has only a sample size of 235 respondents, composed of 174 respondents from Generation Z, 45 respondents from Millennials, and 16 respondents from Generation X. This sample size can limit the reliability of the study. Hence, the result could be more accurate if the number of respondents was increased and equally distributed among the different generations.

In addition, the paper only assesses five e-commerce applications, which can limit the range and generalizability of the study. According to [97], there are more than 50 e-commerce applications used here in the Philippines. Therefore, having more alternatives can enhance the findings and results of the study. Moreover, other capacity factors, such as social media integration, loyalty programs, or advanced security features for online transactions, could contribute to the success of an e-trade website.

Furthermore, the researchers used service quality (SERVQUAL) and multi-criteria data analysis (MCDA) through the use of TOPSIS to investigate and evaluate customer preferences in choosing a trustworthy e-commerce application; however, the researchers had only eleven latent variables to be observed and 72 survey questionnaires, which limit the standardization of the model used. Therefore, future researchers are suggested to include other possible latent variables that have a significant correlation and increase the number of survey questionnaires to attain the standard required by the model and robust generalizability of the study, which will enable future researchers to generate more accurate and reliable results.